Decidual macrophages derived NO downregulates PD-L1 in trophoblasts leading to decreased Treg cells in recurrent miscarriage

Abstract

Introduction: Placental trophoblasts contribute to regulatory T (Treg) function via the programmed cell death-1 (PD-1)/PD-1 ligand 1 (PD-L1) pathway during normal pregnancy. Decreased expression of PD-L1 in trophoblasts was closely associated with Treg deficiency in the development of pregnancy failure. Thus, targeting PD-L1 might be a novel therapy to prevent pregnancy loss. However, the mechanisms for modulating the expression of PD-L1 in trophoblasts are an enigma.

Methods: The proportion of decidual Treg cells, and the profile of decidual macrophages (DMs) sampled from women with normal pregnancy (NP) and recurrent miscarriage (RM) were evaluated by flow cytometry. The expression of Yin and Yang 1 protein (YY1) and PD-L1 in human villous were measured by Immunohistochemistry (IHC), qRT-PCR and western blot. The determination of soluble PD-L1 (sPD-L1) in serum from NP and RM, and trophoblast conditioned media (TCM) was performed by the PD-L1 SimpleStep ELISA kit. Knockdown of YY1 was processed in the human trophoblast derived cell lines, HTR-8 and Bewo, with siYY1 transfection. Peripheral naïve CD4⁺ T cells were isolated from women with NP for the in vitro culture. The percentages of Treg cells differentiated from peripheral naïve CD4⁺ T cells were measured by flow cytometry. The interaction between YY1 and CD274 was proved by CHIP. The expression of inducible nitric oxide synthase (iNOS) in decidua was evaluated by IHC. The level of NO in serum from women with NP and RM was determined by the Griess reagent system. The effects of NO on YY1 were determined by the in vitro culture of HTR-8 cells with the NO donor, SNAP. The in vivo model comprising twelve pregnant mice and underwent different treatment. The percentages of Treg cells in murine uterus were measured by flow cytometry. Similarly, Western blot and IHC were performed to determine the expression of YY1 and PD-L1 in murine placenta.

Results: Decreased expression of YY1 and PD-L1 in trophoblasts and lower proportion of decidual Treg cells were observed in patients with RM. Knockdown of YY1 contributes to a lower expression of YY1 and PD-L1. Soluble PD-L1 in the supernatant from HTR-8 cells was also decreased with siYY1 administration. Lower Treg differentiation was observed in the presence of supernatant from HTR-8 cells treated with siYY1. CHIP analysis revealed that endogenous YY1 directly occupied the promoter region of the CD274 (PD-L1) gene. Accompanied with increased M1 DMs, higher NO was observed in serum sampled from patients with RM. In the presence of Reduced expression of YY1 and PD-L1 was observed in HTR-8 cells with the treatment of SNAP. Furthermore, less Treg differentiation was observed with SNAP treated TCM. Moreover, our in vivo data found that YY1 deficiency was associated with decreased PD-L1, which further resulting in less Treg differentiation and Treg deficiency at the maternal-fetal interface and increased embryo loss.

Discussion: Our work found the modulatory capacity of YY1 on PD-L1 in trophoblasts during early pregnancy. Furthermore, reduced YY1 was supposed resulting from higher levels of NO produced from the M1 DMs in RM.

Keywords: NO; PD-L1; Treg; YY1; decidual macrophage; recurrent miscarriage; trophoblast.